Post on 12-Feb-2022
Investigating a Higher Renewables Portfolio Standard in California
Executive Summary
January 2014
© 2013 Copyright. All Rights Reserved.
Energy and Environmental Economics, Inc.
101 Montgomery Street, Suite 1600
San Francisco, CA 94104
415.391.5100
www.ethree.com
Investigating a Higher Renewables Portfolio Standard in California Executive Summary
January 2014
Table of Contents
Study Sponsors ................................................................................................ 1
Advisory Panel.................................................................................................. 1
Study Team ....................................................................................................... 1
Disclaimer .......................................................................................................... 2
Executive Summary ...................................................................................... 3
Operational Challenges of Achieving a 50% RPS ....................... 9
Potential Integration Solutions ....................................................... 15
Cost and Rate Impacts .................................................................... 19
Effect of Renewable Integration Solutions .................................. 25
Greenhouse Gas Impacts............................................................... 29
Distributed Generation Impacts ..................................................... 30
Next Steps ......................................................................................... 31
Conclusion ......................................................................................... 36
List of Figures
Figure 1: Generation mix calculated for an April day in 2030 with the (a) 33% RPS,
(b) 40% RPS, and (c) 50% RPS Large Solar portfolios showing overgeneration
......................................................................................................................................... 13
Figure 2: Cost differences between RPS portfolios under a range of assumptions;
relative to 2030 33% RPS scenario (2012 cents/kWh) ......................................... 22
Figure 3: Cost impacts of solution cases (assuming 5,000 MW change) under
low and high cost ranges, relative to 2030 33% RPS Scenario (2012
cents/kWh) ..................................................................................................................... 29
List of Tables
Table 1: 2030 Renewable generation by resource type and scenario (in GWh) ..... 7
Table 2: 2030 Overgeneration statistics for the 33%, 40% and 50% RPS Large
Solar Scenarios ............................................................................................................ 14
Table 3: Marginal overgeneration (% of incremental MWh resulting in
overgeneration) by technology for various 2030 RPS scenarios ........................ 15
Table 4: 2030 overgeneration statistics for 50% RPS Large Solar Scenario and
four solution cases ....................................................................................................... 18
Table 5: Average electric rates for each Scenario under a range of input
assumptions (2012 cents/kWh) ................................................................................. 22
Table 6. Percent change in average electric rates of each Scenario relative to 33%
RPS Scenario, under a range of input assumptions (% change in 2012 $) ..... 23
Table 7: 2030 revenue requirement (2012 $ billion) for each Scenario, percentage
change is relative to 33% RPS .................................................................................. 24
Table 8: Cumulative capital investment through 2030, incremental to 33% RPS in
2020, by scenario (2012 $ billion) ............................................................................. 24
Table 9: High and low cost estimates for solution categories modeled in this study
(2012 $) .......................................................................................................................... 27
P a g e | 1 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Study Sponsors
Los Angeles Department of Water and Power (LADWP)
Pacific Gas and Electric Company (PG&E)
Sacramento Municipal Utilities District (SMUD)
San Diego Gas & Electric Company (SDG&E)
Southern California Edison Company (SCE)
Advisory Panel
A four-member independent advisory panel provided input and feedback on the study development and results. The members of the advisory panel include:
Dr. Dan Arvizu – Director and Chief Executive of the National Renewable Energy Laboratory (NREL), Golden, CO
Dr. Severin Borenstein – Director of the University of California Energy Institute and Co-Director of the Energy Institute at Haas School of Business, UC Berkeley
Dr. Susan Tierney – Managing Principal at Analysis Group Inc., Boston, MA
Mr. Stephen Wright – Retired Administrator, Bonneville Power Administration; General Manager, Chelan County Public Utility District
Study Team
Energy and Environmental Economics, Inc. (E3)
ECCO International
DNV KEMA
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 2 |
Disclaimer
The study was conducted by Energy and Environmental Economics, Inc. (E3),
with assistance from ECCO International and DNV KEMA. The study was funded
by LADWP, PG&E, SMUD, SDG&E, and SCE (the utilities). A review committee
consisting of utility and California Independent System Operator (CAISO)
personnel provided technical input on methodology, data and assumptions. The
utilities, the CAISO and Advisory Panel members reviewed and commented on
assumptions, preliminary results, and earlier drafts of this report.
E3 and the consulting team thank the utilities, the CAISO and the Advisory Panel
members for their invaluable input throughout the process of conducting this
analysis. However, all decisions regarding the analysis were made by E3, ECCO
International and DNV KEMA. E3, ECCO International and DNV KEMA are solely
responsible for the contents of this report, and for the data, assumptions,
methodologies, and results described herein.
P a g e | 3 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Executive Summary
This report presents the results of a study of a 50% renewables portfolio
standard (RPS) in California in 2030. The study was funded by the Los Angeles
Department of Water and Power (LADWP), Pacific Gas and Electric Company
(PG&E), the Sacramento Municipal Utilities District (SMUD), San Diego Gas &
Electric Company (SDG&E), and Southern California Edison Company (SCE), (“the
utilities”) to examine the operational challenges and potential consequences of
meeting a higher RPS. This study was conducted by Energy and Environmental
Economics (E3), with assistance from ECCO International. A companion study
was conducted by DMV KEMA examining “smart grid” technologies that may
become available to help alleviate challenges associated with high penetration
of distributed generation. An independent advisory panel of experts from
industry, government and academia was commissioned to review the
reasonableness of the assumptions and to provide input on the study. The
California Independent System Operator (CAISO) also provided input on key
study assumptions.
The utilities asked E3 to study the following questions:
1. What are the operational challenges of integrating sufficient renewable
resources to achieve a 50% RPS in California in 2030?
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 4 |
2. What potential solutions are available to facilitate integration of
variable renewable resources under a 50% RPS?
3. What are the costs and greenhouse gas impacts of achieving a 40% or
50% RPS by 2030 in California?
4. Would an RPS portfolio with significant quantities of distributed
renewable generation be lower-cost than a portfolio of large-scale
generation that requires substantial investments in new transmission
capacity?
5. What are some “least regrets” steps that should be taken prior to—or
in tandem with—adopting a higher RPS?
6. What remaining key issues must be better understood to facilitate
integration of high penetration of renewable energy?
This report describes the analysis that E3 undertook to answer these questions.
E3 studied four scenarios, each of which meets California’s incremental RPS1
needs between 33% and 50% RPS in different ways:
Large Solar Scenario meets a 50% RPS in 2030 by relying mostly on large,
utility-scale solar PV resources, in keeping with current procurement trends.
1 This study assumes that a 50% RPS is defined in the same way as California’s current 33% RPS. The standard requires generation from eligible renewable resources to be equal to or exceed 50% of retail sales. Large hydroelectric resources do not count as eligible renewable resources.
P a g e | 5 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Small Solar Scenario meets a 50% RPS by 2030 by relying mostly on larger,
distributed (1 – 20 MW) ground-mounted solar PV systems. This scenario
also includes some new larger wind and solar.
Rooftop Solar Scenario meets a 50% RPS by 2030 relying in large part on
distributed residential and commercial rooftop solar PV installations, priced
at the cost of installing and maintaining the systems. This scenario also
includes some new larger wind and solar.2
Diverse Scenario meets a 50% RPS in 2030 by relying on a diverse portfolio
of large, utility-scale resources, including some solar thermal with energy
storage and some out-of-state wind.
In addition, the study analyzes two scenarios that serve as reference points
against which to compare the costs and operational challenges of the 50%
scenarios:
33% RPS Scenario meets a 33% RPS in 2030, representing an extension of
the resource portfolio that is already expected to be operational to meet
the state’s current 33% RPS in 2020.
40% RPS Scenario meets a 40% RPS in 2030 by relying mostly on large,
utility-scale solar PV resources.
The geographic scope of the analysis is a combination of the CAISO, LADWP and
the Balancing Area of Northern California (BANC) Balancing Authority Areas. All
scenarios assume that significant investments and upgrades to both the
2 In this scenario, new rooftop PV systems beyond the current net energy metering cap are assumed to count as a renewable generation source towards meeting the state’s RPS. System owners are assumed to be compensated at the cost of installing and maintaining the systems (i.e. rooftop PV is priced at cost in the revenue requirement calculation). No incentives for solar are assumed, nor does the analysis consider any transfers that could occur if system owners were compensated through other mechanisms, e.g., through net energy metering.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 6 |
California electrical grid and the state’s fleet of thermal generators occur
between 2013 and 2030, including the development of a newer, more flexible
fleet of thermal generation. Thus, the results of the 33% RPS Scenario are not
necessarily indicative of the challenges of meeting a 33% RPS in 2020.
Moreover, if these investments are not realized, the operational challenges and
costs of meeting a higher RPS in 2030 might look very different than what is
shown here.
Table 1 shows the mix of renewable resources modeled for each of the
scenarios described above. In addition to the renewable resources added to
meet the RPS target, the study assumes that a total of 7,000 MW of behind-the-
meter solar photovoltaic resources are installed by 2030 under California’s net
energy metering (NEM) policies, enough to meet approximately 5% of total
load. These resources are assumed to reduce retail sales, but they do not count
toward meeting the 2030 RPS.
P a g e | 7 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Table 1: 2030 Renewable generation by resource type and scenario (in GWh)
33% RPS 40% RPS 50% RPS
Large Solar
50% RPS Diverse
50% RPS Small Solar
50% RPS Rooftop
Solar
Utility RPS Procurement
Biogas 2,133 2,133 2,133 4,422 2,133 2,133
Biomass 7,465 7,465 7,465 9,754 7,465 7,465
Geothermal 16,231 16,231 16,231 20,811 16,231 16,231
Hydro 4,525 4,525 4,525 4,525 4,525 4,525
Solar PV - Rooftop 0 943 2,290 2,290 2,290 22,898
Solar PV - Small 6,536 9,365 13,405 13,405 31,724 11,116
Solar PV - Large 22,190 33,504 49,667 29,059 31,349 31,349
Solar Thermal 4,044 4,044 4,044 10,913 4,044 4,044
Wind (In State) 20,789 24,561 29,948 27,659 29,948 29,948
Wind (Out-of-State) 4,985 4,985 4,985 11,854 4,985 4,985
Subtotal, Utility Gen 88,897 107,755 134,693 134,693 134,693 134,693
Customer Renewable Generation
Solar PV – Rooftop, net energy metered
10,467 10,467 10,467 10,467 10,467 10,467
Subtotal, Customer Gen 10,467 10,467 10,467 10,467 10,467 10,467
Total Renewable Generation
Total, All Sources 99,365 118,222 145,160 145,160 145,160 145,160
The study is the first comprehensive effort to assess the technical challenges of
operating the California system at a 50% RPS with high penetration of both wind
and solar energy. This study examines scenarios for California with up to 15% of
electric load served by wind energy, and 28% served by solar energy. This is a
much higher penetration of wind and solar energy than has ever been achieved
anywhere in the world. In Germany, widely known as a world leader in
renewable energy deployment, 21.9% of electricity generation was renewable in
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 8 |
2012, including 7.4% wind and 4.5% solar.3 In Spain, renewable energy
represented 24% of total generation in 2012, including 18% wind and 4% solar.4
Wind served 30% of domestic load in Denmark in 20125; however, Denmark is a
very small system with strong interconnections to the large European grid, and
it frequently sells excess wind energy to its neighbors. Other jurisdictions such
as Norway, New Zealand and British Columbia have served over 90% of electric
load with renewables by counting large hydroelectric resources; these resources
do not count toward California’s RPS.
At the same time, numerous studies have pointed to the need to decarbonize
the electric sector as a key strategy for achieving deep, economy-wide
reductions in greenhouse gas emissions as well as energy security and economic
development benefits.6 To that end, many other jurisdictions have set high
renewables goals. The European Union Renewables Directive mandates that at
least 20% of total energy consumption (including transportation, industrial and
other non-electric fuel uses) come from renewable energy sources by 2020. By
2030, Germany plans to generate 50% of its electricity supply with renewable
sources, including large hydro.7 Finland aims to achieve 38% of final energy
consumption (including transportation, etc.) from renewable energy sources by
3 “Gross electricity generation in Germany from 1990 to 2012 by energy source,” Accessed July 2013. <www.ag-energiebilanzen.de/componenten/download.php?filedata=1357206124.pdf&filename=BRD_Stromerzeugung1990_2012.pdf&mimetype=application/pdf> 4 “Statistical series of the Spanish Electricity System,” Red Electrica, 2013, Accessed August 2013. <http://www.ree.es/ingles/sistema_electrico/series_estadisticas.asp > 5 “Monthly Statistics: Electricity Supply,” Danish Energy Agency, Accessed: August 2013. < http://www.ens.dk/info/tal-kort/statistik-nogletal/manedsstatistik> 6 See http://www.ipcc.ch/pdf/assessment-report/ar4/syr/ar4_syr.pdf, http://www.iea.org/techno/etp/etp10/English.pdf, and http://www.ethree.com/publications/index_2010.php. 7 “Energy Policies of IEA Countries – Germany,” International Energy Agency, 2013 http://www.iea.org/w/bookshop/add.aspx?id=448
P a g e | 9 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
2020 by relying in large part on biomass resources.8 While no country has
served 40 or 50 percent of its load with variable wind and solar resources,
California is not alone in considering potential futures with high renewables.
This study represents an important advancement in understanding the impacts
of achieving a high RPS.
This report is organized as follows. The remainder of this section summarizes
the study’s key findings in response to the six questions posed above. Section 2
describes the analytical approach and key inputs to the study. Section 3
describes the analysis of operational challenges associated with a 50% RPS.
Section 4 introduces a number of potential solutions that may provide lower-
cost ways of integrating renewable resources into the grid. Section 5 presents
the cost and greenhouse gas impacts of achieving a 50% RPS. Section 6
concludes by discussing the results and summarizing the research needs
identified in this study. A series of technical appendices provide details about
the analysis that was conducted.
OPERATIONAL CHALLENGES OF ACHIEVING A 50% RPS
The study utilizes E3’s Renewable Energy Flexibility (REFLEX) Model on ECCO’s
ProMaxLT production simulation platform to investigate the operational and
flexibility requirements associated with a 50% RPS. REFLEX is specifically
designed to investigate renewable integration issues.9 REFLEX performs random
8 “Energy Policies of IEA Countries – Finland,” International Energy Agency, 2013, http://www.iea.org/Textbase/npsum/finland2013SUM.pdf 9 Throughout this report, the term “renewable integration” is used to encompass a range of operational challenges encountered under higher renewable energy penetrations including “overgeneration” of resources,
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 10 |
draws of weather-correlated load, wind, solar and hydro conditions taken from
a very large sample of historical and simulated data. REFLEX thus considers
operational needs associated with high and low load conditions, high and low
hydro conditions, and a range of wind and solar conditions, as well as a broad
distribution of the hourly and sub-hourly operating reserve requirements.
REFLEX runs are presented for four scenarios: (1) the 33% RPS Scenario, (2) the
40% RPS Scenario, (3) the 50% RPS Large Solar Scenario, and (4) the 50% RPS
Diverse Scenario. 10 Additional runs are presented for variations of the 50% RPS
Large Solar Scenario which include the implementation of several potential
renewable integration solutions. This analysis does not attempt to find an
optimal generation mix or set of renewable integration solutions under the 50%
RPS scenarios. Rather, the analysis explores the operational challenges of a 50%
RPS and provides directional information about the potential benefits and cost
savings of the renewable integration solutions.
The largest integration challenge that emerges from the REFLEX runs is
“overgeneration”. Overgeneration occurs when “must-run” generation—non-
dispatchable renewables, combined-heat-and-power (CHP), nuclear generation,
run-of-river hydro and thermal generation that is needed for grid stability—is
greater than loads plus exports. This study finds that overgeneration is
pervasive at RPS levels above 33%, particularly when the renewable portfolio is
whereby electricity supply exceeds demand net of exports, as well as the fuel costs associated with ramping fossil generation to meet load net of renewable generation. 10 REFLEX runs were not conducted for the Small Solar and Rooftop Solar Scenarios. The integration challenges for these scenarios are very similar to those of the Large Solar Scenario; therefore the Large Solar Scenario is used as a proxy for all three high solar scenarios.
P a g e | 11 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
dominated by solar resources. This occurs even after thermal generation is
reduced to the minimum levels necessary to maintain reliable operations.
Figure 1 shows an April day in 2030 under the 33% RPS, 40% RPS, and the 50%
RPS Large Solar Scenarios on which the system experiences both low load
conditions and high solar output. A very small amount of overgeneration is
observed at 33% RPS. The 40% RPS Scenario experiences over 5,000 MW of
overgeneration, while the 50% RPS Large Solar Scenario experiences over
20,000 MW of overgeneration.
Table 2 shows overgeneration statistics for the 33%, 40% and 50% RPS Large
Solar Scenarios. In the 33% RPS scenario, overgeneration occurs during 1.6% of
all hours, amounting to 0.2% of available RPS energy.11 In the 50% RPS Large
Solar case, overgeneration must be mitigated in over 20% of all hours,
amounting to 9% of available RPS energy, and reaches 25,000 MW in the
highest hour. Potential solutions or portfolios of solutions must therefore be
available during large portions of the year and must comprise a large total
capacity.
This study assumes that managed curtailment of renewable generation occurs
whenever total generation exceeds total demand plus export capability. This is
critical to avoid too much energy flowing onto the grid and causing potentially
serious reliability issues. As long as renewable resource output can be curtailed
in the manner assumed here, the study does not find that high penetration of
wind and solar energy results in loss of load. Renewable curtailment is
11 Curtailment as a percentage of available RPS energy is calculated as: overgeneration divided by the amount of renewable energy that is needed to meet a given RPS target.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 12 |
therefore treated as the “default” solution to maintain reliable operations.
However, the study also evaluates additional solutions that would reduce the
quantity of renewable curtailment that is required.
P a g e | 13 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Figure 1: Generation mix calculated for an April day in 2030 with the (a) 33% RPS, (b) 40% RPS, and (c) 50% RPS Large Solar portfolios showing overgeneration
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 14 |
Table 2: 2030 Overgeneration statistics for the 33%, 40% and 50% RPS Large Solar Scenarios
Overgeneration Statistics 33% RPS 40% RPS 50% RPS
Large Solar
Total Overgeneration
GWh/yr. 190 2,000 12,000
% of available RPS energy 0.2% 1.8% 8.9%
Overgeneration frequency
Hours/yr. 140 750 2,000
Percent of hours 1.6% 8.6% 23%
Extreme Overgeneration Events
99th Percentile (MW) 610 5,600 15,000
Maximum Observed (MW) 6,300 14,000 25,000
REFLEX also tests for shortages in “ramping” capability – the ability of the
generation fleet to accommodate large changes in the net load served over one
or more hours.12 While the scenarios evaluated in this study show no instances
of a shortage of ramping capability that would create reliability problems, this
result is driven partly by the assumption that renewable curtailment can be
utilized not just to avoid overgeneration, but also as a tool to manage net load
ramps. In order to ensure reliable operations, REFLEX utilizes “prospective”
curtailment, in which the system operator looks ahead one or more hours,
subject to uncertainty and forecast error, and curtails renewable output in order
to smooth out hourly and multi-hour ramps. This occurs in instances where this
system would otherwise be unable to accommodate the steep upward ramps
from the mid-afternoon “trough” in net load to the evening peak. Planned and
12 Net load is defined as load minus renewable generation.
P a g e | 15 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
carefully-managed curtailment is therefore a critical tool that is used in the
modeling to maintain reliable operations in the face of overgeneration and
ramping challenges caused by the higher RPS.
The quantity of managed renewable energy curtailment increases exponentially
for RPS requirements that move from 40% to 50% RPS. For example, while the
average curtailed RPS energy for the 50% RPS Large Solar Scenario is 9%, the
marginal curtailment—the proportion of the next MWh of renewable resources
added to the portfolio that must be curtailed—is significantly higher: 22-25%
for most renewable resources and 65% for solar PV, as seen in Table 3.
Curtailment amounts to 26% of the RPS energy required to move from a 33% to
50% RPS under the Large Solar Scenario.
Table 3: Marginal overgeneration (% of incremental MWh resulting in overgeneration) by technology for various 2030 RPS scenarios
Technology 33% RPS 40% RPS 50% RPS
Large Solar 50% RPS Diverse
Biomass 2% 9% 23% 15%
Geothermal 2% 9% 23% 15%
Hydro 2% 10% 25% 16%
Solar PV 5% 26% 65% 42%
Wind 2% 10% 22% 15%
POTENTIAL INTEGRATION SOLUTIONS
Implementation of one or more renewable integration solutions may reduce the
cost of achieving a 50% RPS relative to the default renewable curtailment
solution. This study considers the following potential solutions: (1) Enhanced
Regional Coordination; (2) Conventional Demand Response; (3) Advanced
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 16 |
Demand Response; (4) Energy Storage; and, (5) 50% RPS with a more diverse
renewable resource portfolio (the Diverse Scenario). The most valuable
integration solutions are those that can reduce solar-driven overgeneration
during daylight hours when the system experiences low load conditions.
Downward flexibility solutions, including increased exports, flexible loads, and
diurnal energy storage help to mitigate this overgeneration. Alternatively,
procurement of a more diverse portfolio of renewable resources, which includes
less solar and disperses the renewable generation over more hours of the day,
reduces the daytime overgeneration compared to the Large Solar portfolio.
The study evaluates changes in the renewable integration challenges associated
with the 50% RPS Large Solar Scenario from sequentially implementing potential
integration solutions sized at 5,000 MW. The study therefore evaluates the
directional impact of the solution, but does not attempt to identify an optimal
or least-cost set of solutions. Table 4 shows the overgeneration statistics for the
Large Solar Scenario and each of the solutions tested. With only the renewable
curtailment solution implemented, overgeneration is approximately equal to 9%
of the available renewable energy. Integration solutions that provide only
upward flexibility, like conventional demand response, do not significantly
decrease this overgeneration. However, integration solutions that provide
5,000 MW of downward flexibility, such as energy storage, reduce the
overgeneration to between 3% and 4% of the available renewable energy.
P a g e | 17 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Similarly, the diverse portfolio of renewable resources evaluated here reduces
overgeneration to approximately 4% of the available renewable energy.13
This study assesses each of the solutions in isolation, with the aim of indicating
promising directions for further investigation. However, preliminary analysis
suggests that the effects of the various solutions, if implemented together, are
complementary. Because overgeneration increases exponentially at RPS levels
approaching 50%, optimization of the renewable portfolio with a combination
of solutions could substantially reduce the quantity of curtailment required to
meet a 50% RPS. However, avoiding all instances of renewable curtailment may
be cost-prohibitive.
13 This study considers a diverse portfolio consisting of specific quantities of in-state wind, out-of-state wind, solar thermal with energy storage, and other technologies. A different renewable generation mix would result in a different quantity of overgeneration.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 18 |
Table 4: 2030 overgeneration statistics for 50% RPS Large Solar Scenario and four solution cases
Overgeneration Statistics 50% RPS
Large Solar
Enhanced Regional
Coordination
Conventional Demand
Response
Advanced DR or
Energy Storage
Diverse Portfolio
Total Overgeneration
GWh/yr. 12,000 4,700 12,000 5,000 5,400
% of available RPS energy 8.9% 3.4% 8.8% 3.7% 4.0%
Overgeneration
Hours/yr. 2,000 1,000 2,000 1,200 1,300
Percent of hours 23% 12% 23% 14% 15%
Extreme Overgeneration Events
99th Percentile (MW) 15,000 9,900 15,000 9,900 10,000
Maximum Observed (MW) 25,000 20,000 25,000 20,000 19,000
The solution quantities evaluated in these cases are informed by the size of the
overgeneration caused by a 50% RPS, and not by any estimate of the feasibility
or technical potential to achieve each solution. For example, we are not aware
of any detailed studies of the technical potential for pumped storage or
upwardly-flexible loads in California. Battery technologies have not been fully
demonstrated as commercial systems in the types of applications or at the scale
required to address the integration issues identified in this study. Regional
coordination is promising but has progressed slowly over the past decade.
There are likely to be significant challenges to implementing any of these
solutions.
P a g e | 19 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
COST AND RATE IMPACTS
This study calculates the statewide total cost and average retail rate for each of
the 50% RPS scenarios. The 33% and 40% RPS scenarios are shown as a
reference point. The total cost includes the cost of procuring and operating the
renewable and thermal resources considered in this study, the cost of
transmission and distribution system investments needed to deliver the
renewable energy to loads, and non-study-related costs such as the cost of the
existing grid. The costs do not include real-time grid operating requirements
such as maintaining frequency response. The total cost for the study area is
divided by projected retail sales to calculate an average, ¢/kWh rate across all
customer classes.
As a backdrop, the study estimates that the average retail rate in California
could increase from 14.4 ¢/kWh in 2012 to 21.1 ¢/kWh in 2030 (in 2012 dollars),
a 47% increase, before higher levels of RPS beyond the current 33% statute are
taken into consideration. 14 This increase is driven largely by trends outside the
scope of this study, such as the need to replace aging infrastructure, rather than
by RPS policies. Other analysis has estimated that compliance with the current
33% RPS is expected to raise investor owned utility rates by 6-8% between 2011
and 2030; the approximately 40% remaining rate impact expected over this
period would be attributable to other factors.15
14 Throughout the study, all costs are presented in 2012 real dollars unless otherwise noted. 15 This estimate is derived from analysis developed by E3 in the Long-Term Procurement Plan (LTPP) proceeding
http://www.cpuc.ca.gov/NR/rdonlyres/070BF372-82B0-4E2B-90B6-0B7BF85D20E6/0/JointIOULTPP_TrackI_JointIOUTestimony.pdf
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 20 |
Achieving a 40% RPS could lead to an additional increase of 0.7 ¢/kWh, or 3.2%,
over the 33% RPS Scenario under base case assumptions regarding the price of
natural gas, CO2 emissions allowances and renewable energy resources. A 50%
RPS would increase rates by 9 – 23% relative to a 33% RPS under base case
assumptions.
Figure 2, Table 5 and Table 6 below show the average rate increase of each of
the five RPS scenarios compared to the 33% portfolio in 2030. The analysis
reveals several interesting findings:
1. Under a wide range of CO2, natural gas and renewable energy prices (gas
prices from $3-10/MMBtu, CO2 prices from $10-100/metric ton, and a range
of solar PV and wind costs) the higher RPS Scenarios result in an increase in
average electric rates. The rate impacts are expected to be lowest under
the high gas & CO2 price sensitivity with low renewable energy costs.
2. Rate increases are expected to be significantly higher under the 50% RPS
Scenarios than under the 40% RPS Scenario. This is primarily due to the
exponential increase in renewable curtailment as the RPS target increases
towards 50%, requiring a significant “overbuild” of the renewable portfolio
to meet the RPS target.
3. The Diverse Scenario shows a substantially lower rate impact than the more
heavily solar dominated cases, primarily because the diverse portfolio
results in less overgeneration.
4. The rank order on costs between the Scenarios stays the same under all
uncertainty ranges considered. Costs are expected to be highest under the
P a g e | 21 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Rooftop Solar Scenario, followed by the Small Solar, Large Solar and Diverse
Scenarios. The cost differences between these sensitivity results are
reduced when assuming lower solar PV costs than in the base case.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 22 |
Note: RE = renewable energy
Figure 2: Cost differences between RPS portfolios under a range of assumptions; relative to 2030 33% RPS scenario (2012 cents/kWh)
Table 5: Average electric rates for each Scenario under a range of input assumptions (2012 cents/kWh)
33% RPS 40% RPS 50% RPS Large Solar
50% RPS Diverse
50% RPS Small Solar
50% RPS Rooftop
Solar
Average System Rate (2012 cents/kWh)
Base 21.1 21.8 24.1 23.1 24.6 26.1
Low Gas 19.8 20.6 23.0 22.0 23.5 25.0
High Gas 22.9 23.4 25.5 24.4 26.0 27.5
Low CO2 20.5 21.2 23.6 22.5 24.1 25.6
High CO2 22.6 23.1 25.2 24.2 25.7 27.2
Low RE Cost 21.0 21.5 23.1 22.5 23.5 24.5
High RE Cost 21.2 22.1 25.1 23.6 25.7 27.7
Low Gas & CO2, High RE Cost 19.2 20.3 23.5 22.0 24.1 26.1
High Gas & CO2, Low RE Cost 24.2 24.4 25.6 25.0 26.0 27.0
P a g e | 23 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Table 6. Percent change in average electric rates of each Scenario relative to 33% RPS Scenario, under a range of input assumptions (% change in 2012 $)
33% RPS 40% RPS 50% RPS Large Solar
50% RPS Diverse
50% RPS Small Solar
50% RPS Rooftop
Solar
Percentage Change in Average System Rate (relative to 33% RPS)
Base n/a 3.2% 14.0% 9.1% 16.4% 23.4%
Low Gas n/a 4.1% 16.5% 11.3% 19.1% 26.5%
High Gas n/a 2.2% 11.3% 6.6% 13.5% 19.9%
Low CO2 n/a 3.6% 15.2% 10.2% 17.7% 24.9%
High CO2 n/a 2.4% 11.8% 7.1% 14.0% 20.6%
Low RE Cost n/a 2.3% 9.9% 7.1% 11.6% 16.3%
High RE Cost n/a 4.2% 18.1% 11.1% 21.2% 30.5%
Low Gas & CO2, High RE Cost n/a 5.7% 22.3% 14.7% 25.8% 36.0%
High Gas & CO2, Low RE Cost n/a 0.7% 5.8% 3.1% 7.2% 11.3%
The projected cost increases for the higher RPS scenarios are due largely to the
high and increasing cost of renewable integration. While wind and solar costs
are projected to be comparable to the cost of conventional resources on a
levelized cost of energy (LCOE) basis in 2030, overgeneration and other
integration challenges have a substantial impact of the total costs for the 50%
RPS scenarios. Moreover, renewable generation is shown to have very little
resource adequacy benefits beyond 33% RPS due to increased saturation of the
grid with solar energy (see section 2.3 of the full report).
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 24 |
Table 7: 2030 revenue requirement (2012 $ billion) for each Scenario, percentage change is relative to 33% RPS
Revenue Requirement Category 33% RPS 40% RPS
50% RPS Large Solar
50% RPS Diverse
50% RPS Small Solar
50% RPS Rooftop
Solar
CO2 Compliance Cost 3.2 2.9 2.5 2.4 2.5 2.5
Conventional Generation 20.3 19.5 18.7 18.1 18.7 18.6
Renewable Generation 8.2 10.6 17.1 14.8 18.5 22.8
Transmission 6.5 7.1 7.8 7.9 7.4 7.3
Distribution 16.2 16.2 16.3 16.3 16.7 16.5
Misc/Other Costs 2.5 2.5 2.5 2.5 2.5 2.5
Total 56.9 58.8 64.9 62.1 66.3 70.3
Percentage Change n/a 3.2% 14.0% 9.1% 16.4% 23.4%
The total cost of each scenario, in terms of annual revenue requirement in 2030,
is shown in Table 7, while the cumulative capital investment through 2030,
incremental to meeting a 33% RPS in 2020, for each scenario is shown in Table
8.
Table 8: Cumulative capital investment through 2030, incremental to 33% RPS in 2020, by scenario (2012 $ billion)
33% RPS 40% RPS 50% RPS Large Solar
50% RPS Diverse
50% RPS Small solar
50% RPS Rooftop
Solar
New Renewable Generation
9.2 29.5 65.2 61.0 72.0 105.3
New Conventional Generation
11.7 11.2 11.2 11.2 11.2 11.2
New Transmission 2.8 6.6 12.0 15.2 9.3 8.5
New Distribution 0.6 0.9 1.4 1.4 4.2 3.0
Total Capital Investment
24.4 48.1 89.8 88.7 96.6 128.0
P a g e | 25 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
The total increase in annual revenue requirement associated with a 50% RPS in
2030 ranges from $5.2 to $13.3 billion above the 33% RPS scenario; this includes
CO2, fuel and capacity savings in the conventional generation cost category, as
well as increases in renewable procurement costs. The cumulative capital
investment in the 50% RPS Scenario ranges from $64.4 to $103.7 billion above
the 33% RPS scenario, in real 2012 dollars under base case assumptions, before
the implementation of the additional renewable integration solutions that are
investigated in this study.
EFFECT OF RENEWABLE INTEGRATION SOLUTIONS
The cost impacts shown in the tables above incorporate only the “default”
integration solution of renewable energy curtailment. Implementation of one
or more alternative solutions may reduce the cost impacts by enabling a larger
proportion of renewable energy output to be delivered to the grid.
A detailed cost-benefit analysis of these renewable integration solutions is
beyond the scope of this study. In lieu of such an analysis, we provide cost and
rate results under an illustrative range of high and low cost assumptions for the
implementation of 5,000 MW of each of the solutions that are shown to have a
potential renewable integration benefit. Even though the study assumes
significant quantities of each solution (5,000 MW) are implemented, these cases
are not sufficient to fully eliminate the overgeneration challenge.
As noted above, the study does not include an analysis of the optimal level of
integration solutions, nor does it assess the feasibility of procuring or
implementing 5,000 MW of these renewable integration solutions by 2030. The
technical potential to achieve various solutions is unknown, and there are likely
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 26 |
to be significant technical, regulatory and permitting barriers to implementing
solutions at this magnitude.
Table 9 shows the cost ranges assumed for each solution. These assumptions
represent, at a high level, a range of potential costs for each category. In reality,
each category would likely be made up of a number of individual measures or
projects, each of which would have unique costs and benefits. For example, the
energy storage solution case could include a mixture of pumped storage and
other storage technologies such as compressed air energy storage (CAES) or
flow batteries. Nevertheless, this section provides an indication of the extent to
which cost reductions might be achieved through implementation of solutions
in each of these categories.
P a g e | 27 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Table 9: High and low cost estimates for solution categories modeled in this study (2012 $)
Solution Sensitivity Basis Cost Metric
Storage
Low
Pumped hydro cost ($2,230/kW; 30-yr lifetime); Black and Veatch Cost and Performance Data for Power Generation Technologies
16
$375/kW-yr
High Battery cost ($4,300/kW; 15-yr lifetime); Black and Veatch Cost and Performance Data for Power Generation Technologies
$787/kW-yr
Flexible Load
Low Load shift achieved through rate design at no incremental cost
$0/kW-yr
High
Average TRC cost of thermal energy storage ($2,225/kW; 15-yr lifetime); E3 Statewide Joint IOU Study of Permanent Load Shifting
17
$413/kW-yr
Regional Coordination
Low Assume CA receives $50/MWh for exported power
-$50/MWh exported
High Assume CA pays $50/MWh to export incremental power
$50/MWh exported
Figure 3 shows the effect of implementing these solutions, compared to the
33% RPS Scenario. As a benchmark, the 50% RPS Large Solar Scenario, with only
the default renewable curtailment solution, is expected to increase average
16 Study available at: http://bv.com/docs/reports-studies/nrel-cost-report.pdf. 17 Study available at: http://www.ethree.com/public_projects/sce1.php.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 28 |
rates by 3 ¢/kWh, or 14%, relative to the 33% RPS Scenario. The Diverse
Scenario is also shown as an integration solution, along with a point estimate of
its rate impact under base case assumptions. The Diverse Scenario reduces the
average rate by 1 ¢/kWh relative to the Large Solar Scenario.
The Enhanced Regional Coordination and Advanced DR solutions provide cost
savings relative to the Large Solar Scenario, even under the “high” cost range.
The “low” cost range for energy storage, modeled here as 5,000 MW of
relatively low-cost pumped storage, would be expected to reduce the total cost
of achieving the 50% RPS Large Solar Scenario by just over 0.5 ¢/kWh.18 Only
the high-cost battery storage case results in higher costs; however, it should be
noted that the engineering cost estimates shown here do not include site-
specific costs, performance guarantees and other costs that would be incurred
during commercial deployment. All of the solution cases modeled here result in
higher expected rates compared to the 33% RPS Scenario.
18 This study does not assess the feasibility of implementing 5,000 MW of pumped storage in the state by 2030.
P a g e | 29 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Figure 3: Cost impacts of solution cases (assuming 5,000 MW change) under low and high cost ranges, relative to 2030 33% RPS Scenario (2012 cents/kWh)
GREENHOUSE GAS IMPACTS
The 50% RPS Scenarios reduce greenhouse gas (GHG) emissions relative to the
33% RPS Scenario. Increasing the RPS from 33% to 40% reduces GHG emissions
by approximately 6 million metric tons in 2030, while a 50% RPS would reduce
GHG emissions by 14-15 million metric tons relative to a 33% RPS in 2030. The
implied cost of GHG emissions reductions is calculated as the change in total
cost (excluding CO2 compliance costs) divided by the change in GHG emissions
relative to the 33% RPS Scenario.19 The implied carbon abatement cost is
$340/ton for the 40% RPS Scenario, $403/ton for Diverse Scenario, and
$637/ton for the Large Solar Scenario, under the default renewable curtailment
19 This formulation implicitly attributes all of the cost of meeting a higher RPS to GHG emissions reductions. It therefore ignores other potential societal benefits of increased renewable penetration such as reduced emissions of “criteria” pollutants such as NOx and SOx.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 30 |
solution. GHG abatement costs would be reduced if lower-cost, low carbon
solutions to the overgeneration challenge can be implemented.
DISTRIBUTED GENERATION IMPACTS
This study considers two scenarios composed largely of distributed solar PV
generation (DG PV): a Small Solar and a Rooftop Solar Scenario. DG PV is
assessed for its impact on the distribution and transmission systems as well as
for the difference in cost and performance relative to larger systems located in
sunnier areas. The study relies on the methods used in previous studies for the
California Public Utilities Commission (CPUC) to determine both benefits—in the
form of reduced system losses and deferred transmission and distribution
investments—and costs—in the form of distribution system upgrades needed to
accommodate high levels of DG that result in “backflow” from distribution
feeders onto the main grid.
The Small Solar and Rooftop Scenarios are found to be costlier than the Large
Solar and Diverse Scenarios. This is largely due to the difference in cost and
performance assumed for DG PV systems relative to central station systems.
Rooftop systems, in particular, are significantly more expensive to install on a
per-kW basis than larger systems and have lower capacity factors. Rooftop PV
systems tend to have lower capacity factors (partly due to suboptimal tilt and
orientation) relative to larger, ground-mounted systems that can utilize tracking
technologies. DG PV is found to reduce transmission costs relative to larger
systems located in remote areas; however, distribution costs are found to be
higher due to the need for significant investments to accommodate very high
penetration of PV on the distribution system. It should be noted that this study
P a g e | 31 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
does not address issues related to retail rate design or net energy metering.
The DG PV systems are priced at the cost of installing and maintaining the
systems, identical to the treatment of central station PV systems, in the revenue
requirement and average rate calculations.
NEXT STEPS
Although the focus of this study is on grid operations with very high renewable
penetrations in 2030, there are a number of shorter-term “least-regrets”
opportunities that the evidence in this report suggests should be implemented
prior to or in parallel with a higher RPS standard. The four “least regrets”
opportunities identified in this study include:
1. Increase regional coordination. This study shows that increased coordination
between California and its neighbors can facilitate the task of integrating
more renewable resources into the bulk power system at a lower cost.
Although California already depends on its neighbors for imports during
summer peak periods, an increased level of coordination across the West
would include more sharing of flexible resources to support better integration
of the rich endowment of wind energy in the Pacific Northwest and Rocky
Mountains and solar resources in the Desert Southwest.
2. Pursue a diverse portfolio of renewable resources. The study shows that
increasing the diversity of resources in California’s renewable energy portfolio
has the potential to reduce the need for managed curtailment. More diverse
renewable generation profiles can better fit within California’s energy
demand profile. The benefits of developing a diverse portfolio are
complemented by and in many ways tied directly to increased regional
coordination, since the largest benefit is likely to be achieved through
increased geographic diversity across a wide area.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 32 |
3. Implement a long-term, sustainable solution to address overgeneration
before the issue becomes more challenging. A long-term, sustainable
implementation and cost-allocation strategy to manage the potential large
amounts of overgeneration that could result from a higher RPS should be
developed before overgeneration jeopardizes reliability, and before
curtailment impacts financing of new renewable generation projects. A long-
term, sustainable solution must be technically feasible, economically efficient
and implementable in California. It must include a mechanism for ensuring
that renewable developers continue to receive a sufficient return to induce
investment in projects on behalf of California ratepayers.
4. Implement distributed generation solutions. Increased penetration of
distributed generation necessitates a more sustainable, cost-based strategy to
procure distributed generation. This requires a reexamination of retail rate
design and net energy metering policies, as well as implementation of
distribution-level solutions and upgrades, including smart inverters with low-
voltage ride-through capabilities that allow distributed photovoltaic systems
to operate under grid faults.
There are also a number of key areas for future research that are beyond the
scope of this study, but are critical to enable the bulk power systems to
continue to work reliably and efficiently in the future. These include:
The impact of a combined strategy of multiple renewable integration
solutions. This study finds that grid integration solutions will be critical to
achieving a higher RPS at lowest cost. Because each solution has its own
specific costs and benefits, a critical next step is to analyze combinations of
these potential solutions to help develop a more comprehensive, longer-
term grid integration solution to higher RPS.
P a g e | 33 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Research and development for technologies to address overgeneration.
Technology needs to support higher renewable energy penetration and to
address the overgeneration challenge include diurnal energy flexibility and
efficient uses for surplus solar generation during the middle of the day.
Promising technologies include:
Solar thermal with energy storage;
Pumped storage;
Other forms of energy storage including battery storage;
Electric vehicle charging;
Thermal energy storage; and
Flexible loads that can increase energy demand during daylight hours.
Technical potential and implementation of solutions. This study points to
the need for solutions to the renewable integration challenges to be
planned and implemented on the same timeline as, or before, higher
renewable penetration. However, the technical potential to achieve each
solution is unknown at this time. A significant effort is needed to
characterize the technical potential, cost, and implementation challenges
for pumped storage, battery technologies, upwardly-flexible loads, more
diverse renewable resource portfolios, and other potential renewable
integration solutions.
Sub-five minute operations. A better understanding of the sub-five minute
operations, including frequency, inertia and regulation needs, under a
higher RPS is needed. This is particularly pressing in California where
significant changes are planned to the state’s existing thermal generation
fleet, including the retirement of coastal generators utilizing once-through
cooling. Research is needed regarding potential costs and the feasibility
and performance of potential solutions, such as synthetic inertia.
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 34 |
Size of potential export markets for excess energy from California.
California has historically been an importer of significant quantities of
electric energy. Under a 50% RPS, California would have excess energy to
sell during many hours of the year. The extent to which electricity providers
in other regions might be willing to purchase excess energy from California
is unknown. This study assumes that California can export up to 1500 MW
of energy during every hour of the year based on a high-level assessment of
supply and demand conditions in other regions, and shows that higher
levels of exports could significantly reduce the cost of achieving a 50% RPS.
Further research might be able to shed additional light on this question.
Transmission constraints. This study does not include an assessment of
transmission constraints within California, and how those constraints might
impact renewable integration results including reliability, cost and
overgeneration. For example, if a large proportion of the solar energy
resources modeled in this study are located in Southern California,
northbound transmission constraints on Path 15 and Path 26 may result in
significantly higher overgeneration than is indicated in this study.
Challenges may also be more acute within the BANC and LADWP Balancing
Authority Areas, which have limited transfer capability to the CAISO system.
Changing profile of daily energy demand. Daily load shapes are expected
to evolve over time, with increases in residential air conditioning and
electric vehicle loads. This could shift the peak demand period farther into
the evening, potentially exacerbating the overgeneration challenge during
daylight hours.20
20 San Diego Gas & Electric and Sacramento Municipal Utility District are already seeing peak demand occur between 5:30 and 6:00 pm on some days.
P a g e | 35 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
Future business model for thermal generation and market design. This
analysis points toward a fundamental shift in how energy markets are likely
to operate under high penetration of renewable energy. Energy markets
are unlikely to generate sufficient revenues to maintain the flexible fleet of
gas generation that the state will need to integrate high levels of renewable
energy. Moreover, there may be a significant number of hours in which
market prices are negative. New market products for flexibility, inertia,
frequent startups and capacity may be necessary to ensure that the
generation fleet maintains the necessary operating characteristics.
Optimal thermal generation fleet under high RPS. Procurement choices
will need to be made regarding trade-offs between combined-cycle gas
generators, frame and aeroderivative combustion turbines, and other
technologies with newly-important characteristics for renewable
integration, such as low minimum generation levels and high ramp rates.
The flexibility needs of the state’s thermal fleet may also interact with local
air quality regulations, which limit the number of permitted power plant
starts.
Natural gas system impacts and supply. Operating the grid under a higher
RPS may require more flexibility in the natural gas delivery system and
markets. Whether the natural gas delivery system can support the
simultaneous operation of gas-fired generators necessary for renewable
integration is an important area for further research.
Operational challenges of a 40% RPS. The study finds that overgeneration
occurs at 33% RPS and is significant at 40% RPS, but does not evaluate the
impact of renewable integration solutions at a 40% RPS in detail.
Cost-effectiveness of a higher RPS relative to other measures for reducing
GHG emissions. This study indicates that a 50% RPS may be a relatively
high-cost means of reducing GHG emissions (over $300/ton, as compared
Investigating a Higher Renewables Portfolio Standard in California
P a g e | 36 |
to CO2 allowance price forecasts of $30-100/ton). To be sure, there are
many other benefits from higher renewable penetration besides GHG
reduction. Nevertheless, it would be instructive to compare the cost of a
50% RPS with the cost of reducing GHG emissions in other sectors such as
transportation, industry and buildings.
CONCLUSION
This study assesses the operational impacts, challenges, costs, greenhouse gas
reductions, and potential solutions associated with a 50% RPS in California by
2030. The study finds that renewable integration challenges, particularly
overgeneration during daylight hours, are likely to be significant at 50% RPS.
The study indicates that at high penetrations of renewable generation, some
level of renewable resource curtailment is likely to be necessary to avoid
overgeneration and to manage net load ramps. The study also identifies a
number of promising integration solutions that could help to mitigate
overgeneration, including procurement of a diverse portfolio of renewable
resources, increased regional coordination, flexible loads, and energy storage.
Achievement of a higher RPS at least cost to electric customer will likely require
implementation of a portfolio of integration solutions; timely implementation of
these solutions is critical but would likely involve substantial challenges related
to cost, feasibility, and siting. In this study, a 50% RPS is shown to lead to higher
electric rates than a 33% RPS under a wide range of natural gas prices, CO2
allowance prices, and renewable resource costs. The lowest-cost 50% RPS
portfolio modeled here is one with a diversity of renewable resource
technologies. The highest-cost portfolio modeled is one that relies extensively
on rooftop solar photovoltaic systems. This study highlights the need for
additional research in a number of areas, including the need to address sub-five-
P a g e | 37 |
Executive Summary
© 2013 Energy and Environmental Economics, Inc.
minute operational issues, ensure sufficient power system flexibility, and
develop strategies to avoid overgeneration.